Non-myeloablative transplants are being used increasingly for a number of diseases, including Non Hodgkins lymphoma, Hodgkin's disease, myeloma, acute leukemia, chronic lymphoytic leukemia (CLL) and chronic myelogenous leukemia (CML). A reduction in toxicity compared to fully myeloabative allogeneic stem cell transplantation is stimulating the use of this strategy particularly for older patients. Donor engraftment is achieved in the majority of patients, however, a significant number of patients, particularly those with myelodisplastic syndrome (MDS) experience secondary loss of donor engraftment. This can be fatal for some patients due to the prolonged pancytopenia that follows. There are several possibilities for the loss of donor engraftment; 1) donor graft rejection by recipient T cells, 2) late donor graft failure due to insufficient stem cells in the graft, or 3) dominance of recipient stem cells due to competitive repopulation. In allogeneic transplants to support high dose chemotherapy, graft failure/graft rejection occurs early after transplant within the first month, however, for patients that achieve early donor engraftment secondary graft failure/graft rejection is rare and only occurs in less than 5% of patients [1,2]. In contrast, nearly a 100% of patients receiving mini-allogeneic transplants achieve donor engraftment within 2 months of transplant. Graft failure/graft rejection occurs at 2 to 4 months post transplant in approximately 15% of mini-allogeneic recipients. We hypothesize that loss of donor grafts in non-myeloablative stem cell transplant (NST) recipients can occur due to rejection of the donor cells by recipient T cells and/or low numbers of donor stem cells in the graft resulting in secondary graft failure. The aim of this proposal is to develop methods to evaluate graft rejection and determine the mechanisms responsible for late donor graft failure in recipients of NST.